Kawasaki syndrome (KS) is an acute vasculitis of childhood which results in coronary-artery aneurysms or ectasia in 15 to 25% of affected children. The investigators have demonstrated that high- dose intravenous gamma globulin (IVGG) is a safe and effective therapy when administered early in the course of KS. However, the ideal IvGG regimen remains unknown. The proposed study will compare the efficacy of a single large dose of IVGG to that of smaller doses given on each of four consecutive days with regard to 1) severity and duration of systemic inflammation, 2) serum IgG level, and 3) adverse reactions. In addition, the study will explore factors that relate to the occurrence of coronary abnormalities despite IVGG therapy. These aims will be achieved through a multicenter, collaborative, prospective, randomized trial, with randomization stratified by age, sex, and center. To be eligible, subjects must meet the criteria for KS with exclusion of children who present after the tenth day of illness. Enrollment of 100 to 150 subjects per year is anticipated. Children will be randomized to receive either (1) IVGG, 400 mg/kg/day, for four consecutive days, as in the original trial, or (2) IVGG, 2.0 g/kg, as a single infusion. Children in each group will receive aspirin, 100 mg/kg/day, through Day 14 of illness, then 3 to 5 mg/kg/day as a single daily dose. Primary outcome measures include intensity and duration of fever; laboratory variables indicative of acute inflammation (white blood count, C-reactive protein, alpha-l-anti- trypsin, albumin); serum IgG level; adverse reactions; and coronary artery abnormalities. Echocardiograms will be performed with standardized technique and will be interpreted centrally with reviewers blinded to therapy group. Analysis will include comparison of outcomes in the treatment groups using multivariate methods. The relationship of indexes of systemic inflammation and serum IgG level to development of coronary-artery abnormalities will be modeled using logistic regression and discriminant analysis. This study will influence current therapy for KS and guide directions for future research. If the efficacy and safety of gamma globulin given as a single large dose are similar to those of the standard four-day regimen, children with KS will require fewer days of hospitalization or may be treated with out-patient infusion. The identification of factors associated with development of coronary abnormalities despite IVGG treatment will allow modification of dose regimen for children at highest risk.